Thoughts on nailing roof framing directly to old asphalt shingles?

Jul 21, 2026 Last reply: 14 hours ago 4 Replies

I was poking around an old attic recently and saw a roof tie-in that had me scratching my head. The newer rafters were basically sitting right on top of the old shingles. It looked like they just ran a board across the existing roof deck and started nailing the jack rafters right through the asphalt. I've always heard you're supposed to strip it back or at least use sleepers to distribute the weight properly, but this setup has apparently held up for over twenty-five years without any signs of sagging or leaking. It makes me wonder if we overcomplicate things sometimes with modern framing standards. If the original structure is beefy enough to handle the extra weight of the new peak, does it actually matter if the old shingles are still sandwiched in there? Some guys swear by stripping everything down to the deck for a clean connection, while others say it’s just extra work if the base is solid. What do you all think about this kind of over-framing? Is it just a shortcut that happens to work out, or is there a genuine structural reason to avoid it even if it’s been standing for decades? I'm curious to hear how you guys ussually handle these types of additions.


I've run into this plenty of times on older homes in the Midwest where a 1950s ranch got a garage addition back in the 80s. When we did a remodel on a place in Oakhaven last year, we found the exact same thing—2x6 jack rafters nailed right thru three layers of old asphalt shingles into the original 1x8 tongue-and-groove sheathing. The biggest issue I see isn't usally the weight, but how the valley plate sits. If those shingles compress or rot out over time, you lose that tight connection and the nails can start to back out from the vibration and shifting. Personally, I always take the extra hour to strip a three-foot path for the sleeper. I like to get my plate flat against the deck and run a bead of PL Premium or similar adhesive before nailing. It just gives me peace of mind knowing the load is transferring wood-to-wood rather than squishing through grit and fiberglass. Even if the old-timers got away with it for decades using 16d sinkers, modern inspectors around here will flag it the second they see shingle grit under a structural member. It might be over-engineered, but I’d rather sleep well when the heavy snow loads hit.

Leaving shingles in that sandwich is a recipe for a spongy connection that eventually fails. I’ve seen guys try to justify it becasue the shingles are 'allready there,' but those asphalt layers compress over time. When those 2x4 or 2x6 rafters start to settle into the grit, your fastener tension goes right out the window. Every time we do a tie-in for a gabled dormer, I make it a point to snap lines and cut a clean path right down to the plywood or planking.

Getting a flush, wood-on-wood connection with a proper 2x8 or 2x10 valley plate is the only way to ensure the transfer of load is solid. If you leave the old roofing in there, you're essentially building on top of a giant, gritty gasket. I used a Senco nailer on a project last fall where the previous contractor had left the shingles, and half the nails were backing out because they never grabbed the sheathing properly thru the mess. Stripping it back takes twenty minutes with a shingle shovel and a circular saw, and it saves you from a callback when the new ridge starts to dip.

I tried nailing thru old shingles once on a porch roof and the bounce was the worst part. Every time I hit a nail, the whole board woud just spring back becuase of that rubbery asphalt. I ended up grabbing an old circular saw blade I didn't care about and just notched out a channel for my 2x4 plate to sit in. It saved me from stripping the whole slope but got the wood sitting solid. If you don't cut that path, you're fighting the material the whole time you're trying to frame, and your lines will never stay true.

I hear the concerns about 'spongy' connections, but have you ruled out the possibility that the asphalt actualy acts as a gasket in these older builds? If it’s been standing twenty-five years without a single nail backing out or a dip in the ridge, it’s hard to call it a failure. I’ve seen cases where stripping it back to the old deck actually caused more leaks becasue the new fasteners opened up paths thru the original sheathing that the old shingles were previously sealing up. Is it possable we're overthinking the compression aspect when the sheer weight of the roof is doing the heavy lifting?

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required